grsecurity forums

Lately, a couple times a day now, Firefox 31 running on Gentoo hardened 3.15.8 in a QEMU guest, for no apparent reason insofar as I can see is getting closed suddenly, without any pattern of activity. The only explanation I have seen is in the log at "/var/log/messages", which appears to me to indicate that something about grsecurity is terminating the program. Any suggestions as to how I should procede to avert this problem would be welcome. Here is a piece of the log immediately after it occurs, and the most pertinent part of it, if I am not mistaken, is near the end where PAX steps in like so "Aug 12 07:34:07 vsys1p9 kernel: PAX: execution attempt in: <anonymous mapping>...":

Aug 12 07:32:56 vsys1p9 kernel: grsec: denied resource overstep by requesting 25 for RLIMIT_NICE against limit 0 for /usr/bin/pulseaudio[pulseaudio:3316] uid/euid:1000/1000 gid/egid:1001/1001, parent /usr/bin/pulseaudio[pulseaudio:3315] uid/euid:1000/1000 gid/egid:1001/1001
Aug 12 07:32:56 vsys1p9 kernel: grsec: denied resource overstep by requesting 24 for RLIMIT_NICE against limit 0 for /usr/bin/pulseaudio[pulseaudio:3316] uid/euid:1000/1000 gid/egid:1001/1001, parent /usr/bin/pulseaudio[pulseaudio:3315] uid/euid:1000/1000 gid/egid:1001/1001
Aug 12 07:32:56 vsys1p9 kernel: grsec: denied resource overstep by requesting 23 for RLIMIT_NICE against limit 0 for /usr/bin/pulseaudio[pulseaudio:3316] uid/euid:1000/1000 gid/egid:1001/1001, parent /usr/bin/pulseaudio[pulseaudio:3315] uid/euid:1000/1000 gid/egid:1001/1001
Aug 12 07:32:56 vsys1p9 kernel: grsec: more alerts, logging disabled for 9 seconds
Aug 12 07:32:57 vsys1p9 pulseaudio[3316]: [pulseaudio] alsa-util.c: Disabling timer-based scheduling because running inside a VM.
Aug 12 07:32:57 vsys1p9 pulseaudio[3316]: [pulseaudio] alsa-util.c: Disabling timer-based scheduling because running inside a VM.
Aug 12 07:34:07 vsys1p9 kernel: PAX: execution attempt in: <anonymous mapping>, 2bd516c0000-2bd516cd000 2bd516c0000
Aug 12 07:34:07 vsys1p9 kernel: PAX: terminating task: /usr/lib64/firefox/firefox(DOM Worker):3336, uid/euid: 1000/1000, PC: 000002bd516c0008, SP: 000002bd33d24e08
Aug 12 07:34:07 vsys1p9 kernel: PAX: bytes at PC: ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
Aug 12 07:34:07 vsys1p9 kernel: PAX: bytes at SP-8: 000002bd33a93840 000002bd54b011b0 0000000000000182 000002bd32965f80 0000000000000000 fffb82bd30782380 000002bd33d24ea8 000002bd33265e20 000002bd5244ce9e 0000000000000601 fffb82bd30782380
Aug 12 07:34:07 vsys1p9 kernel: grsec: denied resource overstep by requesting 4096 for RLIMIT_CORE against limit 0 for /usr/lib64/firefox/firefox[DOM Worker:3336] uid/euid:1000/1000 gid/egid:1001/1001, parent /bin/bash[bash:3157] uid/euid:1000/1000 gid/egid:1001/1001

how about searching the forum next time?

Fair enough, but I'm not sure that my searching would have been helpful in this case because I think that the person who made that post you found must be a programmer since the links referenced in his comments indicate, if I am not mistaken, a level of programming knowledge well above what I know. So even now having read it, I am not sure how to procede with fixing the problem. Specifically, he concluded:

> "I'm guessing grsec interfere with this and doesn't let their signal handler do its job ?
> The only way I found to make it work is to disable jit or disable PAGEEXEC/SEGMEXEC as pointed out by jbł c ps...

Maybe the end of his post got cut off? Anyway, I am hoping that you can provide me a bit more help to explain what I now need to do to fix it. Can you? Thanks, again.

Whatever the incompatibility was, it disappeared when I upgraded Firefox to version 32.